A cerebral arteriovenous malformation (AVM) is an abnormal tangle of arteries connected directly to veins that shunts blood flow under high pressure and has a propensity to hemorrhage in otherwise healthy young adults. Small AVMs are treated with surgical resection or stereotactic radiation, but neither therapy is both safe and effective for large AVMs. A better understanding of vascular radiobiology may lead to new therapies for these lesions. AVM obliteration after radiation occurs as a result of progressive endothelial depletion and smooth muscle cell proliferation. The endothelial cells appear to be responsible for its radiosensitivity, and the smooth muscle cells appear to be responsible for its occlusion. The mechanism of radiation-induced AVM or arterial occlusion is some combination of smooth muscle cell proliferation, elaboration of secretory protein, and contraction that concentrically narrows the lumen and progressively occludes it. Specific genes and molecular factors that regulate smooth muscle cells have been implicated in this process, namely nitric oxide (NO) and transforming growth factor-beta I (TGF-beta I). TGF01 is a potent stimulator of smooth muscle cell proliferation, and NO is a potent inhibitor. If involved, deletion of these genes from arteries would modulate their response to radiation. Hypothetically, artery without the TGF-beta I gene would have a decreased occlusive response to radiation, whereas artery without the NO synthase gene would have an increased occlusive response. It is hypothesized that NO and TGF-betaI participate in radiationinduced arterial narrowing in a fistula model for AVMs, and that this response can be enhanced by decreasing the inhibitory influence of NO or by increasing the stimulatory influence of TGF-beta I on smooth muscle cells. This research aims to develop the transgenic arteriovenous fistula model, which is an animal model that replicates the angio-architecture and hemodynamics of a simple AVM, and that enables transgenic mouse artery to be inserted at the fistulous site and remain viable over time. The radiation dose and time required to induce occlusive arteriopathy will be established for this model. Finally, the model will be used to examine relative differences in radiation-induced arteriopathy in NOS knock-out, TGF-01 knock-out, and wild-type artery under conditions of fistulous blood flow. This research will determine whether modulating smooth muscle cell proliferation affects radiation-induced arterial occlusion and has therapeutic potential for AVMs. If NO and TGF-01 are involved in the radiation arteriopathy of AVMs, they might be used to enhance the efficacy of conventional stereotactic radiosurgery as part of a gene therapy for high-grade cerebral AVMs.